The AI that knows the difference between a fact, an argument, and a guess.
And tells you which is which, every time.
You’ve used ChatGPT. You’ve used Claude. You’ve used Gemini. They’re brilliant. They’re also confidently wrong, often, in ways you can’t see until it costs you.
SameGROUND is built for the decisions you can’t afford to get wrong.
The problem with every other AI tool
Ask any chatbot a serious question. Should I hire this person? Is this contract sound? Should we launch in Q3? You’ll get a fluent, plausible, authoritative-sounding answer.
You’ll have no idea which parts are based on solid evidence, which parts are reasonable inference, and which parts the model essentially made up to fill the gap.
That’s fine if you’re writing a birthday poem. It’s catastrophic if you’re making a hire, signing a contract, or betting your quarter on a strategy.
How SameGROUND is different
Every SameGROUND diagnostic runs on the SameGROUND framework, a four-layer reasoning system built specifically to fix this problem. The four layers work in order:
Stability.
Before answering, it checks what it actually knows. Sources. Evidence. Drift from the question you asked. If the ground is shaky, it tells you.
Generation.
It builds the answer under constraint, forced to consider opposition, alternatives, and the things you didn’t think to ask.
Expression.
It writes like a person who’s accountable for the answer, not a chatbot trying to please you.
Governance.
Before you see a word of it, the answer is stress-tested by a simulated hostile specialist who tries to tear it apart. What survives is what you read.
The result is an answer where every claim is labelled:
- ESTABLISHED: proven, verifiable, defensible
- ARGUED: reasoned from evidence, with the reasoning shown
- CONJECTURE: a guess, flagged so you know not to bet on it
You finish the diagnostic knowing exactly what you’re relying on.
Four frameworks, one product
The four-layer architecture above runs inside four specialist frameworks, each tuned for a different kind of work. You don’t have to pick. Auto routing reads your input, identifies the right framework, asks any clarifying questions, and runs the diagnostic. If you already know which one you want, you can pick it directly on the run-diagnostic page.
SameGROUND Diagnostic
High-stakes reasoningWhen the question carries material stakes (a hire, a contract, a launch, a market entry, a compliance question), Diagnostic runs the full reasoning protocol. Produces a structured six-section assessment with adversarial review, every claim labelled by epistemic strength.
SameGROUND Creative Production
Dramatic structureFiction, screenplays, story concepts, narrative non-fiction. Beyond prose polish, Creative Production assesses structural choices against thirty governing dramatic laws: tension, stakes, character arcs, the shape of a scene.
SameGROUND Writing Craft
Editorial governanceEditorial work on any written text. Removes AI-generated writing patterns, restores authored voice, lifts prose quality. For writers, editors, and anyone publishing in a world where reader trust depends on whether the writing sounds like a person.
SameGROUND Authored
Long-form non-fictionBook chapters, manuscripts, authored essays, academic work. Long-form prose generation with claims, evidence, and adversarial review of the argument’s load-bearing moves.
What you get back
Every diagnostic produces a document you can forward, file, save, or defend:
- +A clear decision or verdict
- +Every claim labelled by epistemic strength
- +The gaps and risks an expert would raise
- +The opposing case, fairly stated
- +A full audit trail of the reasoning
Show it to your co-founder. Show it to your board. Show it to your lawyer. It holds up.
Where the value shows up
We can’t promise a number on a single diagnostic. The decisions are yours; the stakes vary. What we can show is the kind of cost a careless answer typically carries in each area, so you can judge whether one rigorous second opinion or even creating an artefact from a use case is worth £19.99.
- Typical cost of a bad hire (UK):£30,000 to £50,000
- Compliance gap caught before a regulator:often five figures, sometimes more
- A flawed strategy killed in week two, not month six:a quarter you don’t spend recovering
- A SameGROUND diagnostic:£19.99 one-off, or check out our subscriptions
You don’t need a diagnostic to be right every time. You need it to be rigorous the time it matters, and to tell you clearly which parts of its own answer to trust.
There’s no sales team in the price. The cost reflects the AI processing and the framework itself, not a margin built around a quota carrier. You can see what each tier covers below and pick what fits.
Pricing
Pay As You Go
£19.99 per sessionOne-off purchase. No subscription. One purchase gives you time to complete a full diagnostic session, typically one complex diagnostic or several simpler ones in conversation. Credits never expire. Top up any time.
Starter
£69.99 per monthRoughly one diagnostic per working week. Covers around 5 medium-complex diagnostic sessions per month. Top up with PAYG sessions at £19.99 any time. Cancel any time.
Professional
£299.99 per monthRoughly one diagnostic per working day across your team. Covers around 20 medium-complex diagnostic sessions per month. Top up with PAYG sessions at £19.99 any time. Cancel any time.
Team accounts launching soon. Seat count locks at signup.
All plans include all four SameGROUND frameworks, structured/Markdown/plain output, .docx upload and download, and full session history.
See full pricing details →Try it on a real decision
Don’t take our word for it. Bring SameGROUND a decision you’re actually wrestling with right now. Run one diagnostic. See what it surfaces that you missed.
If it doesn’t earn its place in your toolkit in the first session, you’ll have spent £19.99 on the most useful conversation you’ll have this month.
Want to see the difference between a typical AI answer and a SameGROUND answer first? Look at a few side-by-side comparisons →
SameGROUND. Decisions, governed.